基于火焰图像识别陶瓷辊道窑烧成带工况智能检测方法研究

发布时间：2018-02-24 04:26

本文关键词： 陶瓷辊道窑热电偶火焰图像识别信息融合模糊专家系统　出处：《景德镇陶瓷学院》2015年硕士论文　论文类型：学位论文

【摘要】：陶瓷辊道窑广泛应用于陶瓷制品的生产,它以转动的辊子作为坯体运载的工具。在生产过程中,陶瓷制品被放置在水平密集的耐火辊上,依靠辊子的转动使陶瓷制品从窑头预热带传送到窑尾冷却带,完成陶瓷产品的烧制。陶瓷制品烧制质量的好坏主要取决于陶瓷辊道窑烧成带陶瓷制品烧结状况,而陶瓷制品烧结状况又与烧成带工况密切相关,因此烧成带工况的检测是整个陶瓷辊道窑生产过程的关键环节。由此,研究辊道窑燃烧带工况的有效识别方法具有重要的意义。本文针对目前陶瓷辊道窑温度点检测的缺陷,提出了采用模拟人工观火的火焰图像识别技术与多点温度检测技术相结合识别陶瓷辊道窑烧成带工况的方法,同时设计了基于多传感器的辊道窑燃烧带工况识别的智能检测系统,搭建了仿真平台。主要成果如下：1.提出了基于多传感器的陶瓷辊道窑烧成带温度关键过程数据融合方法。该方法利用热电偶传感器对烧成带进行多点温度采集,并通过信息融合算法融合采集得到的温度数据,从而得到关键过程数据。2.根据陶瓷辊道窑火焰图像的特点,首先选用帧间平滑和中值滤波对火焰图像进行去噪,然后用K-means聚类算法对图像进行聚类实现图像的分割,最后提取得到火焰部分及其r、g、b特征值。3.构建了信息融合模糊专家系统,通过该系统将火焰图像特征与关键过程数据进行融合,得到识别燃烧带工况的有效信息。通过Matlab仿真Fuzzy工具箱进行仿真研究,仿真结果表明该方法的有效性和可行性。
[Abstract]:Ceramic roller kiln is widely used in the production of ceramic products. It uses rotating rollers as the tool for carrying billets. In the process of production, ceramic products are placed on horizontal dense refractory rollers. Depending on the rotation of the roller, the ceramic products are transferred from the kiln head preheating belt to the kiln tail cooling belt to complete the firing of the ceramic products. The quality of the ceramic products firing mainly depends on the sintering condition of the ceramic products fired in the ceramic roller kiln. The sintering condition of ceramic products is closely related to the working condition of firing belt, so the detection of firing belt condition is the key link in the whole production process of ceramic roller kiln. It is of great significance to study the effective identification method of roller kiln combustion zone condition. This paper aims at the defects of the current ceramic roller kiln temperature point detection. This paper presents a method to identify the firing condition of ceramic roller kiln with the combination of flame image recognition technology and multi-point temperature detection technology. At the same time, an intelligent detection system based on multi-sensor for identifying the burning condition of roller kiln is designed. A simulation platform is built. The main results are as follows: 1. A multi-sensor based data fusion method for the critical process of sintering strip temperature in ceramic roller kiln is proposed. The key process data. 2. According to the characteristics of ceramic roller kiln flame image, firstly, the flame image is de-noised by inter-frame smoothing and median filtering. Then K-means clustering algorithm is used to realize image segmentation. Finally, the flame part and its eigenvalue. 3. A fuzzy expert system for information fusion is constructed. Through the fusion of flame image features and key process data, the effective information of identifying combustion condition is obtained, and the simulation results of Matlab simulation Fuzzy toolbox show that the method is effective and feasible.
【学位授予单位】：景德镇陶瓷学院
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ174.6

【相似文献】